See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/321072536

Integrating core and log data in the petrophysical rock type approach to
identify flow units and predict permeability in a carbonate reservoir of
Campos Basin

Conference Paper · March 2017

DOI: 10.1190/sbgf2017-162

CITATIONS READS

0 1,635

2 authors:

Heitor Lichotti Abel Carrasquilla

Universidade Estadual do Norte Fluminense Universidade Estadual do Norte Fluminense
1 PUBLICATION 0 CITATIONS 116 PUBLICATIONS 228 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Abel Carrasquilla on 10 October 2018.

The user has requested enhancement of the downloaded file.

 Integrating core and log data in the petrophysical rock type approach to identify
flow units and predict permeability in a carbonate reservoir of Campos Basin
Heitor Lichotti & Abel Carrasquilla, UENF/LENEP, Macaé – RJ

Copyright 2017, SBGf - Sociedade Brasileira de Geofísica

This paper was prepared for presentation during the 15th International Congress of the
fields of Middle East. These carbonate reservoirs are still
Brazilian Geophysical Society held in Rio de Janeiro, Brazil, 31 July to 3 August, 2017. subjected to an intense diagenetic process, which may
Contents of this paper were reviewed by the Technical Committee of the 15th Interna- include cementation, dissolution, dolomitization, recrystal-
tional Congress of the Brazilian Geophysical Society and do not necessarily represent lization and biogenic action. The lithological complexity
any position of the SBGf, its officers or members. Electronic reproduction or storage of
any part of this paper for commercial purposes without the written consent of the resulting from these processes makes it difficult to char-
Brazilian Geophysical Society is prohibited. acterize these rocks, when compared to siliciclastic ones.
____________________________________________________________________
The Campos Basin is the largest oil producer in the Bra-
Abstract zilian continental margin, currently accounting for more
than 70% of the national production. The oilfields present
This work characterizes the carbonate reservoir of Field in this basin include Albian carbonate reservoirs, with
A, located in the Campos Basin, using wireline logs and averages of porosity and permeability of 25% and 250
laboratory petrophysical data. The marked variation of mD, respectively. Given the complex heterogeneity of
petrophysical properties, a characteristic inherent to car- carbonate reservoirs, normally it exists a low recovery
bonate reservoirs, is a motivating factor for this work, factor, and a difficult correlation between rock properties
which seeks to overcome the difficulties of this evaluation. and geophysical data. The characterization of carbonate
In this sense, this work proposes to divide the reservoir reservoirs through an integrated study between their
region into zones in accord to different petrophysical petrophysical properties and their well logs provides a
characterization models, such as lithofacies descriptions, fundamental understanding of their geometry and their
petrophysical zones, units and flow zones. Dataset from dynamic properties (Bruhn et al., 2003).
two wells in the oilfield was then interpreted, from which
petrophysical parameters such as porosity and permeabil- The integration of geological and petrophysical data al-
ity were derived with a good correlation coefficient with lows the development of a rock - fluid model for the reser-
the core laboratory data. The multiple linear regression, voirs. This type of study identifies the different types of
the Cluster Analysis for Rock Typing and Hydrodynamic rocks that make up the reservoirs, as well as those non-
Flow Unit modules of the Interactive Petrophysics – IP reservoirs. A PRT (Petrophysical Rock Types) is defined
(Senergy, 2016) software were utilized, to evaluate the as a rock interval with singular pore geometry, a certain
Reservoir Quality Index, the Flow Zones and Hydraulic mineralogical composition and with certain specific char-
Flow Units. Once these petrophysical characteristics of acteristics of fluid flow. The fluid rock model relates fluid
the first well (A03) were determined, a blind test and a flow characteristics to certain PRT, using analytical tech-
correlation with the second well (A10) were performed. niques to develop such a model as the porosity and per-
With this methodology, it was possible to predict the po- meability analysis of the samples, plugs gas permeability
tential zones and, thus, to provide information that feed measurements, thin sheet analysis, analysis of electron
geological models that simulate the behavior of the reser- microscope pore structures, capillary pressure by mercury
voir in production. injection, pore throat size distribution and the interpreta-
tion of well logs. The consolidation of the results of these
Introduction various analysis techniques reveals the existence of litho-
In this work, a study of flow zoning of the carbonate res- facies, and it is hoped that similarities between the mod-
ervoir of Field A of the Campos Basin was built up. From els proposed by the various techniques will result (Amae-
this, it was proposed a method for determining permeabil- fule et al., 1993).
ity by combining linear regressions in accord to the flow The classification of the reservoir rock, in accord to
units in which this part of the reservoir is contained. Even- Amaefule et al. (1993), is a synergistic process between
tually, it validates the models developed against experi- geology, petrophysics and reservoir engineering, through
mental data of porosity and permeability of oil reservoir which a reservoir volume is divided into a limited number
rocks. This integration of well log data and laboratory of rock types that have:
petrophysical data can therefore bring forth solutions that a) Different ranges of petrophysical properties (porosity,
lead to the exploration and production of oil in carbonate permeability, water saturation and capillary pressure
reserves and provide a basis for a methodology and characteristics) or distinct correlations between the
workflow for similar reservoirs. fundamental properties.
Most of the hydrocarbon reservoir in the world is located b) Common diagenetic features, including both deposi-
in carbonate rocks. Regarding Lucia (2007), carbonates tional facies and diagenesis, which determined pore
cover about 7% of the Earth's surface, even so, more types and geometries present, and thus the petro-
than 60% of the world's oil reserves and 40% of gas are physical properties.
present in them, of which 70% corresponds to the giant

Fifteenth International Congress of the Brazilian Geophysical Society

 Core and logs to identify flow units and permeability 2
The geological qualification thus allows the PRT, and a slightly smaller pore size, ranging from 2 - 10 mm, and
therefore the petrophysical properties, to be distributed in in this group we still observe some points referring to the
a three-dimensional reservoir model based on a deposi- cementless grainstones, but also identifies the marked
tion model and stratigraphic sequences together with the presence of packstones. In Group 3, blue, are located
understanding of the diagenetic processes controlling Mesopores, with pore sizes ranging from 0.5 - 2 mm, in
Changes in reservoir quality (Al-Tooqi et al., 2014). Pub- this group the dominant lithology is the packtones with
lished examples of PRT fall into two general broad cate- presence of cemented packtones, wackstones and a few
gories. PRT performed by petrophysicists as demonstrat- grainstones. Finally, in Group 4, the micropores, with a
ed by Gomaa et al. (2006) and Jooybari et al. (2010) pore size of less than 0.5 mm, are present all the litholo-
typically shows clear differentiation between groups of gies identified by Petrobras (2012), with a higher percent-
samples in terms of petrophysical parameters (porosity, age of points being related to packstones, cemented
permeability, flow zone indicator and other measurements packstones and wackstones.
related to reservoir performance permeability). However,
these groups tend to be weakly linked to a structural In the HFU module of IP, we first select the well and the
geology, in such a way that the three-dimensional distri- curves that governed the calculations for determining the
bution of the defined PRT is highly ambiguous. On the flow units. For well A03, the selection of the porosity
other hand, the PRT carried out by geologists, like the curve was performed first, as the results of the laboratory
one carried out by Hollis et al. (2010), is commonly based analysis and we do the same for the permeability. Then
on lithofacies groupings combined with one or two diage- we select the number of flow units desired for the model-
netic modifiers. This results in a clear relationship in a ing. Such number was previously determined through the
chronostratigraphic three-dimensional deposition model, SMLP (Stratigraphic Modified Lorenz Plot), not shown in
but with a PRT that may have broadly variable and highly this article. For both wells the number of flow units in
overlapping ranges of petrophysical properties. Skalinski conformity to the SMLP, was 8 flow units. The choice of
and Kenter (2015) recently reviewed the many challenges the margins for determining the clustering of points in
and disruptions involved in PRT, and proposed a com- each flow unit is computed automatically by the IP,
prehensive workflow to address these matters. through cluster analysis. The curves of RQI (RQI =
0.0314(k/e), k being the permeability and e the effective
Method
porosity), Z (pore volume - to - grain volume ratio, Z =
The dataset used in this work includes well logs (gamma e/(1 - e),) and FZI (Flow Zone Indicators, FZI = RQI/Z)
rays, resistivity, density, neutron and sonic) and laborato- are computed automatically by the IP that still calculates
ry basic petrophysical data (porosity and permeability) for the permeability equations for each of the flow units,
wells A3 and A10 drilled in the Field A in the Campos previously determined (Amaefule et al., 1993).
Basin. These wells were selected for their physical prox-
imity and for the similarity of the log data, which indicate For well A03, the FZI values of the plugs range from
that the rocks of both wells have similar characteristics, 0,071 a 259,67. In order to obtain measurements that
composed mainly of a fine grainstone, hairy carbonate best represent the HFU of the reservoir, the plugs that
rock with oncoliths and cemented with calcite. presented near FZI were grouped in the same HFU, ac-
cording to the limits defined with the module cluster anal-
The methodology of this study initially consists of working
ysis as shown in Figure 3a. The IP Software, through the
the available dataset in the commercial software IP (Sen-
module HFU, elaborates, using a linear regression, the
ergy, 2016) and to explore the different interpretation
permeability equations for each of the previously defined
modules that it offers for preliminary petrophysical charac-
flow units, and calculates the correlation factor R 2 for
terization (porosity, permeability, saturation of fluids, etc.).
each of the proposed equations. The equations for each
Afterwards, the PRT concept will be applied, as well as
of the HFU are shown Figure 3b. The graph with the Z vs
the comparison between their different models, and finally
RQI distribution is shown in Figure 3c. For well A10, the
the determination of the best method to estimate the
FZI values of the plugs range from 0.1405 to 21.1688. In
permeability, aiming at the wider characterization of the
order to obtain measurements that best represent the
reservoir. IP software also includes a module called "Hy-
HFU of the reservoir, the plugs that presented near FZI
draulic Flow Units" (HFU) that, from the porosity and
were grouped in the same HFU, according to the limits
permeability data, allows the distribution of the reservoir
defined with the module cluster analysis as shown in
in flow units through the following approaches: Reservoir
Figure 4a. The IP Software, through the HFU module,
Quality Index (RQI) and Winland - Pitman Method (Win-
elaborates, using a linear regression, the permeability
land, 1972; Pitman, 1992). The module allows the selec-
equations for each of the previously defined flow units,
tion of the porosity and permeability curves that will be
and calculates the correlation factor R2 for each of the
taken as input for the determination of flow units.
proposed equations. The equations for each of the HFU
Results are shown Figure 4b. The graph with the Z vs RQI, on
In the porosity v's permeability graph (Figure 1) of the the other hand, distribution is shown in Figure 4c.
R35 curves of Winland (1972) for wells A03 (a) and A10
(b), we can observe a pore grouping that matches the In Figure 5a, we can observe in the track 6 (Permeability
facies identification proposed by Petrobras () in the analy- 1) the permeability estimation by the RQI methodology
sis of Group 1 (green) referring to megapores, with pore calculated on the laboratory porosity, and in track 7 Per-
sizes <10μm, there is a predominance of cemented meability by the RQI methodology calculated on the sonic
grains. In Group 2 (yellow), we have the macropores with porosity, in black we have the results of the laboratory
measurements on the plugs. It is also observed that these

Fifteenth International Congress of the Brazilian Geophysical Society

 Heitor Lichotti & Abel Carrasquilla 3
________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

permeability curves, both the permeability from the labor- Acknowledgments

atory porosity and the permeability from the sonic log,
present both excellent behavior when compared to the We thank UENF/LENEP for the computational infrastruc-
permeability measured in the laboratory which indicates ture and Petrobras by the dataset.
that the model of flow units, it was very consistent in the
References
case studied. The permeability estimation based on the
laboratory porosity data, adjusted in accord to the RQI
model, presented very good results, with a correlation − Amaefule, J.; Altunbay, M.; Tiab, D.; Kersey, D. &
factor R2 = 89.99 %. The permeability estimation based Keelan, D. 1993. Enhanced reservoir description:
on the sonic porosity data, adjusted in accord to the RQI using core and log data to identify hydraulic flow units
model, also presented good results, with a correlation and predict permeability in uncored intervals/wells.
factor R2 = 88.55 %. Figure 5b is a plot generated in the SPE Annual Technical Conference and Exhibition,
IP, with the permeability curves generated by the two Housto, p. 205 - 220.
models mentioned above, compared with the laboratory − Al-Tooqi, S.; Ehrenberg, S.; Al-Habsi, N.; Al-Shukaili,
measurements and with the facies models, flow zones M. 2014. Reservoir rock typing of upper Shu Aiba li-
and flow units for the Well A10. As occurred in well A03, mestones, northwestern Oman. Petroleum Geoscien-
in well A10 the permeability model from the equations for ce, v. 20, n. 4, p. 339 -352.
the flow units again is very consistent. − Bruhn,C.; Gomes, J.; Lucchese Jr., C. & Johann, P.
2003. Campos Basin: reservoir characterization and
In an effort to better understand the carbonate reservoir management - historical overview and future challen-
through the two wells analyzed, we looked for ways to ges. Offshore Technology Conference. OTC 15220.
correlate the identified zones in well A03 and A10. Fore- − Gomes, J; Ribeiro, M.; Strohmenger, C.; Naghban, S.
most, the flow zones determined for each well were ana- & Kalam, M. 2008. Carbonate reservoir rock typing -
lyzed in the reservoir region in a try to find a correspond- the link between geology and scal. SPE Abu Dhabi In-
ing pattern (Figure 6a). In a second moment, the corre- ternational Petroleum Exhibition and Conference, Pa-
spondences between the conventional logs are exam- per 118284-MS.
ined. We discovered that the gamma, density, neutron − Gomaa, N.; Azer, S.; Ouzzane, D.; Saif, O.; Okuyiga,
and sonic logs presented the same pattern of behavior in M.; Allen, D.; Rose, D.; Ramamoorthy, R. & Bize, E.
the two wells studied, confirming the correlations between 2006. Case study of permeability, vug quantification,
the previously correlated flow units (Figure 6b). and rock typing in a complex carbonate. SPE Annual
Conclusions Technical Conference and Exhibition, Santo Antonio,
paper 102888- MS.
The permeabilities calculated by modeling the values of − Hollis, C.; Vahrenkamp, V.; Tull, S.; Mookerjee, A.;
the laboratory porosities and the sonic log presented both Taberner, C. & Huang, Y. 2010. Pore system charac-
excellent correlations with the permeability measured in terisation in heterogeneous carbonates: an alternative
the laboratory. This shows the efficiency of the reservoir approach to widely - used rock - typing methodologi-
division method in flow units, which allowed the elabora- es. Marine and Petroleum Geology, v. 27, n. 4, p. 772
tion of different equations to calculate the permeability, as - 793.
a function of the porosity, for each of the HFU previously − Jooybari, H.; Movazi, G. & Jaberi, R. 2010. A new
defined. These individualized equations were later com- approach for rock typing used in one of the Iranian
bined in a single equation that interpolates values accord- carbonate reservoir (a case study). SPE International
ing to the RQI, to the depth, resulting in a correlation Oil and Gas Conference and Exhibition, Beijing, paper
factor of more than 90% in both analyzed porosities. 131915-MS.
Permeability from sonic porosity could then be applied to − Lucia, J. 2007. Carbonate reservoir characterization:
both wells to predict permeability in the untested regions an integrated approach. Springer - Verlag Berlin Heil-
from the RQI analysis. This fact gives indications of its delberg, 343 p.
applicability and extrapolation of these results in other − Petrobras. 2012. UENF / Petrobras Cooperation
wells in the field studied. From the correlation of the two Agreement.
studied wells, we can deduce that the zones and flow − Pittman, E. 1992, Relationship of porosity to permea-
units present in A3 are all identified also in A10. However, bility to various parameters derived from mercury in-
in the latter well they are displaced to a higher region, jection - capillary pressure curves for sandstone.
about 30m high, which Suggests A10 was drilled in a AAPG Bulletin, v. 76, n. 2, p. 191 - 198.
region on a structural top, in a more central position of the − Skalinski, M. & Kenter, J. 2015. Carbonate pe-
reservoir. The potentially more productive zones indicated trophysical rock typing: integrating geological attribu-
for the two wells have all predominance of lithofacies of tes and petrophysical properties while linking with
uncemented grains with pore throat radius classified as dynamic behaviour. Special Publications, Geological
mega or macropores. This gives good permoporous Society, London, v. 406, n. 1, p. 229–259.
characteristics to these zones, which is compatible with − Senergy. 2015. Interactive Petrophysics Software.
the RQI analysis, elaborated for the same sites. The User´s Manual.
integration of petrophysical, geophysical and geological
− Winland, H. 1972. Oil accumulation in response to
data helped to better infer the properties of the reservoir.
pore size changes, Weyburn field, Saskatchewan:
It was observed that the location of the wells influences
Amoco Production Company. Report F72-G-25, 20 p.
the quality of the reservoir rock that it crosses and the
final productivity of each well.

Fifteenth International Congress of the Brazilian Geophysical Society

 Core and logs to identify flow units and permeability 4

Figure 1. Lithofacies (track 2) and basic logs of wells A03 and A10: gamma ray and transit time logs
(track 3), resistivity logs (track 4) and density and neutron porosity (track 5) for wells A03 e A10.

(a)

Figure 2. Distribution of pore sizes with Winland

(1972) R35 curves for wells A03 (a) and A10 (b):
green ( 10μm), yellow (2 - 10 m), blue light
(0.5 - 2 m) and pink (<0.5 m), corresponding
to, respectively, mega (grainstones), macro (not
(b) cimented grainstones and packstones), meso
(cimented packstones, wackstones and few
grainstones) and micro (packstones, cimentes
packstones and wackstones) pores.

Fifteenth International Congress of the Brazilian Geophysical Society

 Heitor Lichotti & Abel Carrasquilla 5
________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

(a) (b)

(c)

Figure 3. Well A03: (a) Flow units and ranges of corresponding IZF values. (B) Equations for permeability
based on laboratory porosity - UFs. (C) Distribution z versus RQI (calculated for laboratory porosity).

(a) (b)

(c)

Figure 4. Well A10: (a) Flow units and ranges of corresponding IZF values. (B) Equations for permeability
based on laboratory porosity - UFs. (C) Distribution z versus RQI (calculated for laboratory porosity).

Fifteenth International Congress of the Brazilian Geophysical Society

 Core and logs to identify flow units and permeability 6

(a) (b)
(a) (b)
Well A03 Well A10

(c)

Figure 5. Lithofacies (track 2), Lorenz curve (track 3), flow zones (track 4), flow units (track 5), permeability curves for wells
A03 (a) and A10 (b), compared with laboratory measurements (tracks 6 and 7).
Figure 3. Well A03: (a) Flow units and ranges of corresponding IZF values. (B) Equations for permeability
based on laboratory porosity - UFs. (C) Distribution z versus RQI (calculated for laboratory porosity).

(a) (b)
(a) (b)

(c)

Figure 4. Well A10: (a) Flow units and ranges of corresponding IZF values. (B) Equations for permeability
based
Figure on laboratorybetween
6. Comparative porositythe
- UFs.
flow(C) Distribution
zones z versus
(a) and between theRQI (calculated
basic forof
well logs (b) laboratory
the wells porosity).
A03 and A10.

Fifteenth International Congress of the Brazilian Geophysical Society

View publication stats